A machine learning framework for sport result prediction

Bunker, Rory; Thabtah, Fadi

doi:10.1016/j.aci.2017.09.005

Cited by 204 publications

(131 citation statements)

References 21 publications

Supporting

Mentioning

129

Contrasting

Unclassified

Order By: Relevance

“…To corroborate this view from data science experts, we also checked that CRISP-DM is still a very common methodology for data mining applications. For instance, just focussing on the past four years, we can find a large number of conventional studies applying or slightly adapting the CRISP-DM methodology to many different domains: healthcare [18], [19], [20], [21], signal processing [22], engineering [23], [24], education [25], [26], [27], [28], [29], logistics [30] production [31], [32], sensors and wearable applications [33], tourism [34], warfare [35], sports [36] and law [37]. However, things have evolved in the business application of data mining since CRISP-DM was published.…”

Section: Crisp-dm and Related Process Modelsmentioning

confidence: 99%

CRISP-DM Twenty Years Later: From Data Mining Processes to Data Science Trajectories

Martínez-Plumed

Contreras-Ochando

Ferri

et al. 2021

IEEE Trans. Knowl. Data Eng.

240

101

View full text Add to dashboard Cite

Section: Crisp-dm and Related Process Modelsmentioning

confidence: 99%

CRISP-DM Twenty Years Later: From Data Mining Processes to Data Science Trajectories

Martínez-Plumed

Contreras-Ochando

Ferri

et al. 2021

IEEE Trans. Knowl. Data Eng.

240

101

View full text Add to dashboard Cite

“…It will push the instructive framework to screen the understudies execution logically. [8] Because of the particular idea of match-related highlights to various games, results crosswise over various investigations in this application can for the most part not be looked at straightforwardly. In spite of the expanding utilization of ML models for sport expectation, increasingly precise models are required.…”

Section: Related Work Analysismentioning

confidence: 99%

Student Future Prediction System Under Filtering Mechanism

2019

IJITEE

View full text Add to dashboard Cite

Information is progressively being utilized to improve regular day to day existence less demanding and. Applications, for example, holding up time estimation, traffic expectation, and stopping look are genuine instances of how information from various sources can be utilized to encourage our day by day life.In this period of computerization, instruction has additionally patched up itself and isn't constrained to old address strategy. The ordinary mission is on to discover better approaches to make it increasingly successful and effective for understudies. These days, loads of information is gathered in instructive databases, yet it remains unutilized. So as to get required advantages from such major information, amazing assets are required. Information mining is a developing integral asset for investigation and forecast. In this investigation, we consider an under-used information source: college ID cards. Such cards are utilized on numerous grounds to buy nourishment, enable access to various territories, and even gauge participation in classes. In this article, we use information from our college to investigate use of the college wellness focus and fabricate an indicator for future visit volume. The work makes a few commitments: it exhibits the extravagance of the information source, demonstrates how the information can be utilized to improve understudy administrations, finds fascinating patterns and conduct, and fills in as a contextual analysis outlining the whole information science process. One objective of this article is to show the information science process. As far as information accumulation, two arrangements of information-timestamp information from card swipes at the Student Recreation Centre (SRC) and client profile information-were gathered from our Management of University. The gathered information was cleaned and further prepared. At that point, exploratory information investigation was performed to find fascinating examples. General understudy practice patterns were found from the timestamp dataset. These incorporate yearly/month to month/every day frequencies of understudy visits to the SRC and pinnacle hours amid multi day.

show abstract

“…Although there are a number of methods, a structured approach would be useful in obtaining best results in the case of sport result prediction problems. The proposed approach focuses on team sports [1]. The approach is based on the CRISP-DM Framework as shown in Fig.1…”

Section: Proposed Approachmentioning

confidence: 99%

Integrating Feature Selection and Multiclass Classification for Sport Result Prediction

Isaac*,

2019

IJITEE

View full text Add to dashboard Cite

Machine learning (ML) has become the most predominant methodology that shows good results in the classification and prediction domains. Predictive systems are being employed to predict events and its results in almost every walk of life. The field of prediction in sports is gaining importance as there is a huge community of betters and sports fans. Moreover team owners and club managers are struggling for Machine learning models that could be used for formulating strategies to win matches. Numerous factors such as results of previous matches, indicators of player performance and opponent information are required to build these models. This paper provides an analysis of such key models focusing on application of machine learning algorithms to sport result prediction. The results obtained helped us to elucidate the best combination of feature selection and classification algorithms that render maximum accuracy in sport result prediction.

show abstract

A machine learning framework for sport result prediction

Cited by 204 publications

References 21 publications

CRISP-DM Twenty Years Later: From Data Mining Processes to Data Science Trajectories

CRISP-DM Twenty Years Later: From Data Mining Processes to Data Science Trajectories

Student Future Prediction System Under Filtering Mechanism

Integrating Feature Selection and Multiclass Classification for Sport Result Prediction

Contact Info

Product

Resources

About